Home   |  Products   |  Contact


The 2015 THUMS European Users' Meeting

JSOL is delighted to announce The 2015 THUMS European Users' Meeting.
THUMS, the Total Human Model for Safety for use with LS-DYNA® is being rapidly adopted by users worldwide.
We invite you to join us and share in THUMS technical information.


Time title
10:00- Opening
10:10- Development and application of THUMS at Toyota
Dr. Tsuyoshi Yasuki, Toyota Motor Corporation
Toyota Motor Corporation (TMC) and Toyota Central Research and Development Laboratory (TCRDL) have developed Total Human Model for Safety (THUMS) since 1999, and have distributed it to OEMs and academic institutions in the world. Current THUMS Version 4 has a familiar of AM95, AM50, and AF05.THUMS has been validated with the test results in the literature by not only TMC/TCRDL, but also academic institutions to confirm its bio-fidelity. Coming new features of THUMS are child model 3YO, 6YO, and 10YO, and muscle activation of THUMS. How the new features are utilized for the research at Toyota is introduced.
10:40- Development and Validation of THUMS Version 5 and Its Application for Vehicle Safety Research
Dr. Masami Iwamoto, Toyota Central R&D Labs., Inc.
Toyota Central R&D Labs. Inc. (TCRDL) have developed a human body FE model called THUMS (Total HUman Model for Safety) with Toyota Motor Corporation (TMC) to simulate motions and injuries of human body during impacts. We have been developing the THUMS with active muscles since 2007, because there are some reports that muscle activity of occupants in pre-crash might affect occupant motions and injury outcomes during crash. TCRDL and TMC have a plan to release new version of THUMS with multiple 1D active muscles, which we call THUMS Version 5, in near future. New features of THUMS Version 5 with its validation results and its application to frontal and side impact simulations are introduced.
11:10- Break
11:20- Latest developments and applications of THUMS-D within Daimler AG
Mr. Christian Mayer, DAIMLER AG
THUMS-D has been applied in several applications over the last years within the vehicle safety division of Daimler. Within these use cases often also biomechanic validity respectively material models were improved or developed in collaboration with our scientific partners. The paper gives an overview of latest development and applications of the THUMS-D family. The creation of a 5% female model is shown and also the approach to implement material models to represent the elderly population is discussed. New application fields are presented with the use of HBM in the production area and the simulation of a human-robot interaction. Finally the progress of an active THUMS-D respectively the modelling of muscles is shown. Need for further research is discussed for all application fields. Within the conclusions also a strategic outlook concerning the future use of HBM within Daimler is given.
11:50- THUMS User Community . Challenges in Validating THUMS in Different Crash Codes
Ms. Therese Fuchs, University of Munich
In recent years Finite-Element (FE) Human Body Models (HBMs) considerably gained in importance not only for the evaluation of occupants' passive safety systems, but also in the field of pedestrian protection. A multiscale validation of the Human Models is necessary to safeguard credible crash simulation results. However, results are also highly dependent on the code used for the simulation. The THUMS User Community aims to develop an approach which is capable of delivering credible harmonised crash simulation results even when using different crash codes. THUMS User Community is a project of LMU in cooperation with Adam Opel AG, AUDI AG, Autoliv, BMW AG, Daimler AG, Porsche AG, Toyota Motor Cooperation and Volkswagen Aktiengesellschaft. Currently, THUMS is used by the project partners in 3 different codes, Abaqus, LS-DYNA and VPS. In a first step, harmonised THUMS models were developed in the mentioned 3 codes. After, a multi-stage validation catalogue was developed representing different levels of validation. As a first level, a basic validation check was conducted on the harmonised models to ensure a robust simulation and an improved comparability between the crash codes. A second-stage validation catalogue aims to safeguard the
biomechanical validity of the models. Several challenges arose when conducting the validation work like the insufficient documentation of the reference paper, multiple differing possibilities to model the setup of the load cases and finally challenges caused by the use of different crash codes.
12:20- Lunch
13:30- Rib fracture prediction with THUMS in frontal impacts
Dr. Damien Subit, Arts et Metiers ParisTech, France (University of Virginia, USA)
The number of rib fractures is commonly used as a proxy to evaluate the severity of the loading applied to the ribcage during a vehicle crash. Indeed, the risk of injuries such as pneumothorax and lung perforations increases with the increasing number of rib fractures. Therefore, it is important for computational human body models to be capable of properly predicting rib fractures. In this study, THUMS v4 was used to simulate the frontal sled tests performed at the University of Virginia (also known as 'gold standard') : post-mortem human subjects (PMHS) were placed on a rigid seat, restrained with a 3-pt belt and a knee bolster, and subjected to a 40 km/h impact. A parametric analysis was designed to evaluate the effect of THUMS initial posture and of the failure strain in the rib cortical bone on the number of rib fractures. The initial posture of THUMS was varied to replicate the posture documented in the PMHS tests. The simulation results show that the initial (pre-impact) posture could play an important role in the number of rib fractures, as the reaction forces with the seat and the belt, and the subject kinematics were found to be significantly altered when the THUMS posture was modified. The results of this study indicate that the subject global impact response is an important factor for the prediction of rib fractures that could be as significant as the value of the failure strain in the rib cortical bone.
14:00- A generic Geometry Modification Method for Positioning and Scaling of Human Models
Dr. Dirk Fressmann, DYNAmore GmbH
Human body modeling has become very important in the automotive industry within the last couple of years. One issue with human models is that most of the well-known models are only available in standardized sizes and postures, e.g. 50/95%ile male, 5%ile female or 6 year-old child among others. However, real human bodies vary in size, geometry and stiffness distribution. Therefore, worst-case scenarios are often aspired, where the size and shape of the human body is chosen, such that the model behavior in a crash test scenario leads to extreme responses in terms of injury criteria or vehicle interaction. These worst-case geometries of the human body are usually obtained by using different body sizes. However the shape of the body also has a tremendous influence and is hardly accounted for. Examples for different body shapes are for instance skinny or obese shapes or changes due to ageing effects. Since the worst-case geometries are not known in advance, various simulations have to be performed for each crash test scenario to study the behavior of different body shapes. This however requires a fast and reliable method for modifying the human body model, according to specific parameters to modify e.g. the shape of the thorax or the abdomen, the size of the extremities or the posture of the human body. This presentation addresses a generic geometry modification tool which can for instance be used to adapt the size and shape of a human body model, based on a non-linear, multi-step interpolation approach on the nodal coordinate level. The background will be presented and examples using the THUMS human body model will be shown.
14:30- Simulation based THUMS positioning on J-SEATdesigner
Dr. Noriyo Ichinose, JSOL Corporation
THUMS is one of the most detailed human FE model and has quite complex structure of the spine, elbow joint, knee joint, and so on. This precise modeling helps us to investigate injury of human under impact loading but makes model positioning more difficult. Unlike FE dummy model, THUMS has no kinematic joint definition (*Constraint_Joint_XXX) at moving parts. This means that almost pre-processor cannot move THUMS parts to desired position with kinematic calculation. One good approach for THUMS positioning is simulation based positioning. However the simulation requires "know-how" to get proper deformation of moving parts. J-SEATdesigner is one of sled analysis system implemented on Oasys/Primer which is developed by Arup. To build the sled model easily, automatic dummy model positioning with kinematic calculation is implemented on J-SEATdesigner. Now we are extending the positioning feature for THUMS on J-SEATdesigner and investigating proper setting for positioning simulation. In this presentation, THUMS positioning feature on J-SEATdesigner will be introduced and show some investigation results about simulation setting for THUMS positioning.
15:00- Closing address

Note: The program is subject to be changed.

Contact us :

World wide distributors

Top of Page